Management of Academic Workload Allocation Using Multi-Objective Genetic Algorithm

Enforcing fairness policies for academic workload distribution and achieving staff satisfaction is of great importance in academic institutions. The amount of effort spent by instructors in teaching individual courses is not only measured by the contact teaching hours with students in classes. Teaching efforts include both in-class and out-of-class activities such as course preparation, teaching, marking exams, marking assignments, and supervising projects. In this paper, the fairness of workload allocation is treated as an optimization problem. We propose a two-dimensional and multi-objective implementation of the Genetic algorithm. The problem is solved using two optimization criteria:1) maximize the fairness workload allocation concerning the actual effort and time spent on the teaching and learning process, and 2) maximize a developed fair eligibility score for instructor and course assignments in workload schedules. The eligibility score is a combined metric which consists of additional factors that may affect the workload allocation decisions such instructor preferences, head of department recommendations, and the level of instructor's expertise in the course. The workload problem is represented using two-dimensional matrices. The experiments are conducted on a real dataset consisting of 32 courses and 10 instructors. The overall performance of the algorithm is measured based on the fitness value and running time of the program. The results on the real dataset show that the proposed algorithm solves the problem efficiently in 395 seconds runtime. The proposed algorithm achieves fair allocation of workload and fair eligibility score with 3.2 and 13 standard deviations respectively. The average eligibility score achieved is 61%.